1
0
Fork 0
mirror of https://git.sr.ht/~seirdy/seirdy.one synced 2024-11-27 14:12:09 +00:00

Compare commits

...

4 commits

Author SHA1 Message Date
Rohan Kumar
7e6b979fdc
Fucking quotes 2023-06-15 11:49:18 -07:00
Rohan Kumar
e9a2af4a2b
Fix connectivity check for ipv4-only builds 2023-06-15 11:47:57 -07:00
Rohan Kumar
103532b3f9
syndicate 2023-06-15 11:41:05 -07:00
Rohan Kumar
b2f62b577a
New note: irrevocable consent 2023-06-15 11:39:21 -07:00
3 changed files with 32 additions and 4 deletions

View file

@ -28,8 +28,5 @@ tasks:
cd seirdy.one
export PATH=~/bin:$PATH
# quick health + ip check
curl -6 --proto "=https" --proto-default https --http2 -siSL --tlsv1.3 --cert-status 'seirdy.one/ip'
echo
curl -4 --proto "=https" --proto-default https --http2 -sSL --tlsv1.3 --cert-status 'seirdy.one/ip'
echo
sh scripts/connectivity-check.sh
bmake deploy-prod deploy-onion

View file

@ -0,0 +1,14 @@
---
title: "Irrevocable consent"
date: 2023-06-15T11:39:21-07:00
syndicatedCopies:
- title: 'The Fediverse'
url: 'https://pleroma.envs.net/notice/AWixDJSQJPt824CMZE'
---
When I use my legal rights to request an organization to delete my data...
1. Do they also un-teach their ML models everything they learned from my data?
2. Can they infer the missing information from that model without storing it?
3. Do they also have data that isn't currently attributed to me, but can be attributed to me in the future?
These are rhetorical questions because you can probably guess the answers. Simple data deletion is a cop-out from the impossible task of un-learning. Consent isn't as meaningful if it isn't fully revocable.

View file

@ -0,0 +1,17 @@
#!/bin/sh
set -e -u
alias curl_cmd='curl --proto "=https" --proto-default https --http2 -siSL --tlsv1.3 --cert-status'
ipv6_success=1
ipv4_success=1
curl_cmd -6 'seirdy.one/ip' || ipv6_success=0
echo
curl_cmd -4 'seirdy.one/ip' || ipv4_success=0
echo
if [ "$ipv6_success" = 0 ] && [ "$ipv4_success" = 0 ]; then
echo "Failed to connect"
exit 1
fi